October 4 2023 PSYC 217 L11: Special Research Designs Learning Goals Identify different special designs that are neither experimental nor correlational Compare and contrast quasi-experimental and correlational designs Explain how 5 threats to internal validity affect the interpretation of results Recognize and explain examples of threats to internal validity Compare and contrast different special designs in terms of threats to internal validity Program evaluation research on programs that are proposed and implemented to achieve a positive effect on a group of people ? Needs Assessment -> what programs need to be addressed in a target pop ? EX what do homeless people need most? o Data: derived from surveys, interviews, existing archival data > Program theory -> program needs to be designed based on causes of problem and best wayt oa address them o Cld involve researchers, service providers, program clients to ensure program addresses populations needs > Process evaluation -> during the program, researchers monitor whether its reaching target pop, attracting enough clients, and staff is providing the planned services > Outcome eval -> assess the impact, are goals being achieved, do participants like it > Efficiency assessment -> benefits worth the costs? Is ther ea more effienient way to do this? Special Designs Single Case Design > ABA Reversal design > ABAB Reversal design > Multiple baseline design Quasi-experiments > One-group posttest only > One-group posttest pretest > Nonequivalent control group > Nonequivalent control group pretest-posttest Multiple repeated measures (in textbook) > Interrupted time series
> Control series Developmental Designs > Cross-sectional > Longitudinal > sequential Reversal Designs ? Greg, ingest pancakes, ingest power up drink within pancake mix, o Measure talking speed o Few days of no drink, w drink, no drink Ā· Establishes baseline first few days > Single reversal ABA design is susceptible to one important problem o One test phase = don't know what else happened in that test phase > Multiple reversals (aba, ababa) minimize the chance of this problem affecting your results > Single case designs - in general - suffer from a lack of generalisability > Threats to Validity -> ABA reversal design = bad for all threats to validity > A (baseline period) -> B (treatment period) -> A (baseline period) o behaviour be observed and measured during the baseline control period (A), again during the treatment period (B), and also during a second baseline control period (A) after removing the experimental treatment. Multiple baseline -> researcher looks for changes in behaviour after a manipulation is introduced under multiple circumstances - Beavhiour of several particopants measured offer time Quasi-Experimental Designs > No randomization > Uses pre-existing groups (or allows participants to sort themselves into groups) > Randomization is not possible or realistic ? Cross-cultural studies, to do with identity One-group Posttest Only Design > Participant group -> treatment (don't have to have treatment) -> posttest > Smoking cessation program offered -> students who sign up -> smoking cessation program -> smoking frequency (1 pack/week) o Problem = you have no